B. Pasternak, "Processing Imprecise and Structural Distorted Line Drawings by an Adaptable Drawing Interpretation Kernel," Proc. DA$ '94, 349-363.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....systems express these manipulations in a concise and easy to understand high level form. 2. Software architectures A variety of software organizations have been used in diagram recognition systems, as is briefly reviewed here. Efforts are underway to provide reusable code for recognition systems [28] [32] 34] The blackboard architecture is a general and flexible framework for combining diverse knowledge sources; applications include recognition of engineering drawings [35] mail pieces [36] and text [31] as well as construction of a drawing interpretation kernel [28] Knowledge sources ....

.... recognition systems [28] 32] 34] The blackboard architecture is a general and flexible framework for combining diverse knowledge sources; applications include recognition of engineering drawings [35] mail pieces [36] and text [31] as well as construction of a drawing interpretation kernel [28]. Knowledge sources communicate via a blackboard data structure. The blackboard represents multiple, conflicting recognition hypotheses, divided into levels of abstraction (e.g. raw image, thresholded image, labeled image, textline, and block [36] The blackboard contents trigger invocation of ....

B. Pasternak, "Processing Imprecise and Structural Distorted Line Drawings by an Adaptable Drawing Interpretation Kernel," Proc. DA$ '94, 349-363.

.... techniques have been developed for a variety of notations, including engineering drawings [21] 22] 33] 35] 72] mathematics notation [1] 10] 17] 25] 28] 71] music notation [4] 9] 23] 36] chemical structure diagrams [46] circuit diagrams [14] 37] 51] and line drawings [13] [52] [62] There is need for a classification or categorization of diagrams, to provide a vocabular for discussing the input domain of a diagram recognition system. Existing work in this area is surveyed in [8] The level of interp retation The following levels of interpretation occur in diagram ....

....are summarized here. So far, no recognition kernels are in widespread use. The development of widely used recognition kernels would certainly help the field of diagram recognition to mature. Pasternak describes an adaptable drawing interpretation kernel that is implemented as a blackboard system [52]. Declarative geometrical constraints are applied to iteratively combine graphical objects into higher level objects. The kernel supplies generally applicable operations such as thresholding, line finding, and vectorization. Domain specific knowledge bases must be added to configure the system for ....

B. Pasternak, "Processing Imprecise and Structural Distorted Line Drawings by an Adaptable Drawing Interpretation Kernel," Proc. IAPR Workshop on Document Analysis Systems, Kaiserslautern, Germany, Oct. 1994, 349-363. 14

....in a number of applications and a great deal of work has been performed to provide robust techniques to interpret them. Both on and off line applications have been considered, including computer graphical user interfaces [7, 26] and conversions for CAD input or for tidying plans or schematics [3, 22, 4, 23, 13, 1]. These images contain an inherent degree of distortion and inaccuracy that the techniques must be able to accommodate, however, the distortion produced by the ROCF is beyond the capabilities of these systems. Neural networks [20, 30] and other adaptive learning techniques [16, 11, 12, 17] have ....

B. Pasternak. Processing imprecise and structural distorted line drawings by an adaptable drawing interpretation kernel. International Association for Pattern Recognition systems, pages 318 -- 37, 1994.

.... examples of work in this area are the ANON system from the University of Sheffield [11] where field knowledge is represented using frames and the low level primitives are assembled into higher level features using a parser, the work done in Germany around the SPRITE project and the WIZ prototype [6, 19], where knowledge sources guide the vectorization, the dimension analysis and the post processing for conversion to 3 D CAD, and our own work with the CELESSTIN system, where expert modules in a blackboard based multi expert system cooperate to extract functional information from a view [24, ....

B. Pasternak. Processing Imprecise and Structural Distorted Line Drawings by an Adaptable Drawing Interpretation Kernel. In Proceedings of IAPR Workshop on Document Analysis Systems, Kaiserslautern (Germany), pages 349--365, 1994.

....Generally applicable operations include thresholding, line finding, and vectorization. Such general knowledge can form the kernel of a diagram recognition system, with the addition of domain specific knowledge bases to configure the system for recognition of particular diagrammatic notations [Past94]. 4. FRAMEWORKS FOR DIAGRAM RECOGNITION Various system organizations and various processing techniques have been proposed for general diagram recognition. The appendix provides a brief review of various recognition frameworks, and how they have been applied to diagram recognition and document ....

.... and grayscale levels) the character level, the linguistic level (divided into word, phrase, and sentence levels) and the textual structure level (divided into paragraph and document levels) Pasternak uses blackboard driven building of aggregates in an adaptable drawinginterpretation kernel [Past94]; declarative geometrical constraints are applied to iteratively combine graphical objects into higher level objects. Knowledge sources can be organized in various ways. For example, Novak and Bulko suggest five types of knowledge sources for symbol arrangement analysis [NoBu93] which group ....

B. Pasternak, "Processing Imprecise and Structural Distorted Line Drawings by and Adaptable Drawing Interpretation Kernel," Proc. IAPR Workshop on Document Analysis Systems, Kaiserslautern, Germany, Oct. 1994, pp. 349-363.

No context found.

B. Pasternak. Processing Imprecise and Structural Distorted Line Drawings by an Adaptable Drawing Interpretation Kernel. Proceeding of the IAPR-Workshop on Document Analysis Systems, pages 349 - 365, Kaiserslautern, Germany, 1994

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC